xref: /netbsd-src/usr.bin/make/str.c (revision ce2c90c7c172d95d2402a5b3d96d8f8e6d138a21)
1 /*	$NetBSD: str.c,v 1.25 2006/08/11 19:11:00 christos Exp $	*/
2 
3 /*-
4  * Copyright (c) 1988, 1989, 1990, 1993
5  *	The Regents of the University of California.  All rights reserved.
6  *
7  * This code is derived from software contributed to Berkeley by
8  * Adam de Boor.
9  *
10  * Redistribution and use in source and binary forms, with or without
11  * modification, are permitted provided that the following conditions
12  * are met:
13  * 1. Redistributions of source code must retain the above copyright
14  *    notice, this list of conditions and the following disclaimer.
15  * 2. Redistributions in binary form must reproduce the above copyright
16  *    notice, this list of conditions and the following disclaimer in the
17  *    documentation and/or other materials provided with the distribution.
18  * 3. Neither the name of the University nor the names of its contributors
19  *    may be used to endorse or promote products derived from this software
20  *    without specific prior written permission.
21  *
22  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
23  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
24  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
25  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
26  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
27  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
28  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
29  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
30  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
31  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32  * SUCH DAMAGE.
33  */
34 
35 /*-
36  * Copyright (c) 1989 by Berkeley Softworks
37  * All rights reserved.
38  *
39  * This code is derived from software contributed to Berkeley by
40  * Adam de Boor.
41  *
42  * Redistribution and use in source and binary forms, with or without
43  * modification, are permitted provided that the following conditions
44  * are met:
45  * 1. Redistributions of source code must retain the above copyright
46  *    notice, this list of conditions and the following disclaimer.
47  * 2. Redistributions in binary form must reproduce the above copyright
48  *    notice, this list of conditions and the following disclaimer in the
49  *    documentation and/or other materials provided with the distribution.
50  * 3. All advertising materials mentioning features or use of this software
51  *    must display the following acknowledgement:
52  *	This product includes software developed by the University of
53  *	California, Berkeley and its contributors.
54  * 4. Neither the name of the University nor the names of its contributors
55  *    may be used to endorse or promote products derived from this software
56  *    without specific prior written permission.
57  *
58  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
59  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
60  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
61  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
62  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
63  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
64  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
65  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
66  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
67  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
68  * SUCH DAMAGE.
69  */
70 
71 #ifndef MAKE_NATIVE
72 static char rcsid[] = "$NetBSD: str.c,v 1.25 2006/08/11 19:11:00 christos Exp $";
73 #else
74 #include <sys/cdefs.h>
75 #ifndef lint
76 #if 0
77 static char     sccsid[] = "@(#)str.c	5.8 (Berkeley) 6/1/90";
78 #else
79 __RCSID("$NetBSD: str.c,v 1.25 2006/08/11 19:11:00 christos Exp $");
80 #endif
81 #endif				/* not lint */
82 #endif
83 
84 #include "make.h"
85 
86 /*-
87  * str_concat --
88  *	concatenate the two strings, inserting a space or slash between them,
89  *	freeing them if requested.
90  *
91  * returns --
92  *	the resulting string in allocated space.
93  */
94 char *
95 str_concat(const char *s1, const char *s2, int flags)
96 {
97 	int len1, len2;
98 	char *result;
99 
100 	/* get the length of both strings */
101 	len1 = strlen(s1);
102 	len2 = strlen(s2);
103 
104 	/* allocate length plus separator plus EOS */
105 	result = emalloc((u_int)(len1 + len2 + 2));
106 
107 	/* copy first string into place */
108 	memcpy(result, s1, len1);
109 
110 	/* add separator character */
111 	if (flags & STR_ADDSPACE) {
112 		result[len1] = ' ';
113 		++len1;
114 	} else if (flags & STR_ADDSLASH) {
115 		result[len1] = '/';
116 		++len1;
117 	}
118 
119 	/* copy second string plus EOS into place */
120 	memcpy(result + len1, s2, len2 + 1);
121 
122 	return(result);
123 }
124 
125 /*-
126  * brk_string --
127  *	Fracture a string into an array of words (as delineated by tabs or
128  *	spaces) taking quotation marks into account.  Leading tabs/spaces
129  *	are ignored.
130  *
131  * returns --
132  *	Pointer to the array of pointers to the words.
133  */
134 char **
135 brk_string(const char *str, int *store_argc, Boolean expand, char **buffer)
136 {
137 	int argc, ch;
138 	char inquote, *start, *t;
139 	const char *p;
140 	int len;
141 	int argmax = 50, curlen = 0;
142     	char **argv = emalloc((argmax + 1) * sizeof(char *));
143 
144 	/* skip leading space chars. */
145 	for (; *str == ' ' || *str == '\t'; ++str)
146 		continue;
147 
148 	/* allocate room for a copy of the string */
149 	if ((len = strlen(str) + 1) > curlen)
150 		*buffer = emalloc(curlen = len);
151 
152 	/*
153 	 * copy the string; at the same time, parse backslashes,
154 	 * quotes and build the argument list.
155 	 */
156 	argc = 0;
157 	inquote = '\0';
158 	for (p = str, start = t = *buffer;; ++p) {
159 		switch(ch = *p) {
160 		case '"':
161 		case '\'':
162 			if (inquote) {
163 				if (inquote == ch)
164 					inquote = '\0';
165 				else
166 					break;
167 			}
168 			else {
169 				inquote = (char) ch;
170 				/* Don't miss "" or '' */
171 				if (start == NULL && p[1] == inquote) {
172 					start = t + 1;
173 					p++;
174 					inquote = '\0';
175 					break;
176 				}
177 			}
178 			if (!expand) {
179 				if (!start)
180 					start = t;
181 				*t++ = ch;
182 			}
183 			continue;
184 		case ' ':
185 		case '\t':
186 		case '\n':
187 			if (inquote)
188 				break;
189 			if (!start)
190 				continue;
191 			/* FALLTHROUGH */
192 		case '\0':
193 			/*
194 			 * end of a token -- make sure there's enough argv
195 			 * space and save off a pointer.
196 			 */
197 			if (!start)
198 			    goto done;
199 
200 			*t++ = '\0';
201 			if (argc == argmax) {
202 				argmax *= 2;		/* ramp up fast */
203 				argv = (char **)erealloc(argv,
204 				    (argmax + 1) * sizeof(char *));
205 			}
206 			argv[argc++] = start;
207 			start = NULL;
208 			if (ch == '\n' || ch == '\0')
209 				goto done;
210 			continue;
211 		case '\\':
212 			if (!expand) {
213 				if (!start)
214 					start = t;
215 				*t++ = '\\';
216 				ch = *++p;
217 				break;
218 			}
219 
220 			switch (ch = *++p) {
221 			case '\0':
222 			case '\n':
223 				/* hmmm; fix it up as best we can */
224 				ch = '\\';
225 				--p;
226 				break;
227 			case 'b':
228 				ch = '\b';
229 				break;
230 			case 'f':
231 				ch = '\f';
232 				break;
233 			case 'n':
234 				ch = '\n';
235 				break;
236 			case 'r':
237 				ch = '\r';
238 				break;
239 			case 't':
240 				ch = '\t';
241 				break;
242 			}
243 			break;
244 		}
245 		if (!start)
246 			start = t;
247 		*t++ = (char) ch;
248 	}
249 done:	argv[argc] = NULL;
250 	*store_argc = argc;
251 	return(argv);
252 }
253 
254 /*
255  * Str_FindSubstring -- See if a string contains a particular substring.
256  *
257  * Input:
258  *	string		String to search.
259  *	substring	Substring to find in string.
260  *
261  * Results: If string contains substring, the return value is the location of
262  * the first matching instance of substring in string.  If string doesn't
263  * contain substring, the return value is NULL.  Matching is done on an exact
264  * character-for-character basis with no wildcards or special characters.
265  *
266  * Side effects: None.
267  */
268 char *
269 Str_FindSubstring(const char *string, const char *substring)
270 {
271 	const char *a, *b;
272 
273 	/*
274 	 * First scan quickly through the two strings looking for a single-
275 	 * character match.  When it's found, then compare the rest of the
276 	 * substring.
277 	 */
278 
279 	for (b = substring; *string != 0; string += 1) {
280 		if (*string != *b)
281 			continue;
282 		a = string;
283 		for (;;) {
284 			if (*b == 0)
285 				return UNCONST(string);
286 			if (*a++ != *b++)
287 				break;
288 		}
289 		b = substring;
290 	}
291 	return NULL;
292 }
293 
294 /*
295  * Str_Match --
296  *
297  * See if a particular string matches a particular pattern.
298  *
299  * Results: Non-zero is returned if string matches pattern, 0 otherwise. The
300  * matching operation permits the following special characters in the
301  * pattern: *?\[] (see the man page for details on what these mean).
302  *
303  * Side effects: None.
304  */
305 int
306 Str_Match(const char *string, const char *pattern)
307 {
308 	char c2;
309 
310 	for (;;) {
311 		/*
312 		 * See if we're at the end of both the pattern and the
313 		 * string. If, we succeeded.  If we're at the end of the
314 		 * pattern but not at the end of the string, we failed.
315 		 */
316 		if (*pattern == 0)
317 			return(!*string);
318 		if (*string == 0 && *pattern != '*')
319 			return(0);
320 		/*
321 		 * Check for a "*" as the next pattern character.  It matches
322 		 * any substring.  We handle this by calling ourselves
323 		 * recursively for each postfix of string, until either we
324 		 * match or we reach the end of the string.
325 		 */
326 		if (*pattern == '*') {
327 			pattern += 1;
328 			if (*pattern == 0)
329 				return(1);
330 			while (*string != 0) {
331 				if (Str_Match(string, pattern))
332 					return(1);
333 				++string;
334 			}
335 			return(0);
336 		}
337 		/*
338 		 * Check for a "?" as the next pattern character.  It matches
339 		 * any single character.
340 		 */
341 		if (*pattern == '?')
342 			goto thisCharOK;
343 		/*
344 		 * Check for a "[" as the next pattern character.  It is
345 		 * followed by a list of characters that are acceptable, or
346 		 * by a range (two characters separated by "-").
347 		 */
348 		if (*pattern == '[') {
349 			++pattern;
350 			for (;;) {
351 				if ((*pattern == ']') || (*pattern == 0))
352 					return(0);
353 				if (*pattern == *string)
354 					break;
355 				if (pattern[1] == '-') {
356 					c2 = pattern[2];
357 					if (c2 == 0)
358 						return(0);
359 					if ((*pattern <= *string) &&
360 					    (c2 >= *string))
361 						break;
362 					if ((*pattern >= *string) &&
363 					    (c2 <= *string))
364 						break;
365 					pattern += 2;
366 				}
367 				++pattern;
368 			}
369 			while ((*pattern != ']') && (*pattern != 0))
370 				++pattern;
371 			goto thisCharOK;
372 		}
373 		/*
374 		 * If the next pattern character is '/', just strip off the
375 		 * '/' so we do exact matching on the character that follows.
376 		 */
377 		if (*pattern == '\\') {
378 			++pattern;
379 			if (*pattern == 0)
380 				return(0);
381 		}
382 		/*
383 		 * There's no special character.  Just make sure that the
384 		 * next characters of each string match.
385 		 */
386 		if (*pattern != *string)
387 			return(0);
388 thisCharOK:	++pattern;
389 		++string;
390 	}
391 }
392 
393 
394 /*-
395  *-----------------------------------------------------------------------
396  * Str_SYSVMatch --
397  *	Check word against pattern for a match (% is wild),
398  *
399  * Input:
400  *	word		Word to examine
401  *	pattern		Pattern to examine against
402  *	len		Number of characters to substitute
403  *
404  * Results:
405  *	Returns the beginning position of a match or null. The number
406  *	of characters matched is returned in len.
407  *
408  * Side Effects:
409  *	None
410  *
411  *-----------------------------------------------------------------------
412  */
413 char *
414 Str_SYSVMatch(const char *word, const char *pattern, int *len)
415 {
416     const char *p = pattern;
417     const char *w = word;
418     const char *m;
419 
420     if (*p == '\0') {
421 	/* Null pattern is the whole string */
422 	*len = strlen(w);
423 	return UNCONST(w);
424     }
425 
426     if ((m = strchr(p, '%')) != NULL) {
427 	/* check that the prefix matches */
428 	for (; p != m && *w && *w == *p; w++, p++)
429 	     continue;
430 
431 	if (p != m)
432 	    return NULL;	/* No match */
433 
434 	if (*++p == '\0') {
435 	    /* No more pattern, return the rest of the string */
436 	    *len = strlen(w);
437 	    return UNCONST(w);
438 	}
439     }
440 
441     m = w;
442 
443     /* Find a matching tail */
444     do
445 	if (strcmp(p, w) == 0) {
446 	    *len = w - m;
447 	    return UNCONST(m);
448 	}
449     while (*w++ != '\0');
450 
451     return NULL;
452 }
453 
454 
455 /*-
456  *-----------------------------------------------------------------------
457  * Str_SYSVSubst --
458  *	Substitute '%' on the pattern with len characters from src.
459  *	If the pattern does not contain a '%' prepend len characters
460  *	from src.
461  *
462  * Results:
463  *	None
464  *
465  * Side Effects:
466  *	Places result on buf
467  *
468  *-----------------------------------------------------------------------
469  */
470 void
471 Str_SYSVSubst(Buffer buf, char *pat, char *src, int len)
472 {
473     char *m;
474 
475     if ((m = strchr(pat, '%')) != NULL) {
476 	/* Copy the prefix */
477 	Buf_AddBytes(buf, m - pat, (Byte *)pat);
478 	/* skip the % */
479 	pat = m + 1;
480     }
481 
482     /* Copy the pattern */
483     Buf_AddBytes(buf, len, (Byte *)src);
484 
485     /* append the rest */
486     Buf_AddBytes(buf, strlen(pat), (Byte *)pat);
487 }
488